ABSTRACT Treatment for drug dependence is undergoing a paradigmatic shift that recognizes the limitations of acute care interventions. There is a growing consensus that drug dependence represents a chronic disease process that requires ongoing attention beyond formal treatment. Long-term professional treatment is prohibitive in part because reduction of cost continues to be an important goal among private and public funders. There is therefore an urgent need to identify non-professional interventions that can help sustain recovery after treatment, or in some cases, serve as an alternative to it. This application builds upon our ongoing and previous research assessing longitudinal outcomes of individuals residing in sober living houses (SLHs). The essential characteristics of SLHs include an alcohol and drug free living environment and social support for abstinence. SLHs are primarily financed through resident fees and residents in most houses can stay as long as they wish. There are over 800 SLHs in California alone, yet there are few studies assessing outcomes. Our research on SLHs shows that residents make significant, sustained (18-month) improvement on a variety of outcomes, including alcohol and drug use, arrests, employment, and other areas of functioning. This application proposes to study contextual factors neglected by our research to date, including how physical, social, and organizational characteristics among a heterogeneous sample of SLHs and the neighborhoods where they are located affect outcome. Factors assessed will include the social environment of houses, organizational and leadership characteristics, and architecture (e.g., size, spatial layout, proximity to neighbors, safety, use of rooms, and upkeep of the facility). We also examine the influence of neighborhood characteristics such as alcohol outlet density and proximity, activity spaces, crime, socioeconomic status, density and proximity of 12-step meetings, density and proximity of treatment services, and interactions between neighbors and SLH residents. Focus groups with residents and house managers will inform implementation of study procedures and understanding of quantitative findings. Procedures will include recruiting a sample of 600 individuals entering 40 purposively selected SLHs in Los Angeles County. Primary outcomes will include substance use, HIV risk, arrests, and employment. Multilevel modeling will parse out relative influences of individual, house, and neighborhood factors on outcome at 6 and 12 months. Results will provide new data on house and neighborhood characteristics that are critical for describing evidence-based SLHs and developing new standards based on scientific findings. Current recovery house organizations, such as the National Alliance of Recovery Residences (NARR), have standards primarily based on practical experience and anecdotal reports. There is therefore an urgent need for research that can help them develop evidence-based house environments and neighborhood locations. Dissemination of findings will target researchers, SLH operators, and SLH organizations involved in the creation of standards.